Optimal sub-graphical models
نویسندگان
چکیده
We investigate the problem of reducing the complexity of a graphical model (G,PG) by finding a subgraph H of G, chosen from a class of subgraphs H, such that H is optimal with respect to KL-divergence. We do this by first defining a decomposition tree representation for G, which is closely related to the junction-tree representation for G. We then give an algorithm which uses this representation to compute the optimal H ∈ H. Gavril [2] and Tarjan [3] have used graph separation properties to solve several combinatorial optimization problems when the size of the minimal separators in the graph is bounded. We present an extension of this technique which applies to some important choices of H even when the size of the minimal separators of G are arbitrarily large. In particular, this applies to problems such as finding an optimal subgraphical model over a (k − 1)-tree of a graphical model over a k-tree (for arbitrary k) and selecting an optimal subgraphical model with (a constant) d fewer edges with respect to KL-divergence can be solved in time polynomial in |V (G)| using this formulation.
منابع مشابه
MaxSAT-Based Cutting Planes for Learning Graphical Models
A way of implementing domain-specific cutting planes in branch-andcut based Mixed-Integer Programming (MIP) solvers is through solving so-called sub-IPs, solutions of which correspond to the actual cuts. We consider the suitability of using Maximum satisfiability solvers instead of MIP for solving subIPs. As a case study, we focus on the problem of learning optimal graphical models, namely, Bay...
متن کاملDynamic Programming Bipartite Belief Propagation For Hyper Graph Matching
Hyper graph matching problems have drawn attention recently due to their ability to embed higher order relations between nodes. In this paper, we formulate hyper graph matching problems as constrained MAP inference problems in graphical models. Whereas previous discrete approaches introduce several global correspondence vectors, we introduce only one global correspondence vector, but several lo...
متن کاملExtraction of Geometric Features with Functional Models
In this contribution, we show how to encode in a convenient and compact way information from images with so called functional graphical models. Those models permit to represent in a coherent mathematical framework features such as interest points , simply connected curves such as line and conic segments, connected regions of homogeneous properties, and compound features such as lines and line p...
متن کاملAn optimization algorithm for free-form surface partitioning based on weighted gaussian image
Partitioning free-form surfaces into sub-patches and finding optimal representative normal for each patch to maximize a global objective function is an important two-level operation in diverse industrial applications. In this paper, by solving a maximum hemispherical partitioning problem raised from a weighted Gaussian image, an optimization algorithm is proposed to partition a free-form surfac...
متن کاملConditional Gaussian Graphical Models for Multi-output Regression of Neuroimaging Data
Pattern recognition has shown considerable promise for automated diagnosis and for predicting outcome in clinical neuroimaging studies. Many of these decision problems can most naturally be framed as multi-output learning problems but nearly all studies to date have adopted sub-optimal analytical approaches, modeling different outputs independently. In this work, we apply a dual elastic net reg...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2004